Transgenerational epigenetic inheritance

Epigenetic categories

Four general categories of epigenetic modification are known:

1. self-sustaining metabolic loops, in which a mRNA or protein product of a gene stimulates transcription of the gene; e.g. Wor1 gene in Candida albicans
2. structural templating in which structures are replicated using a template or scaffold structure on the parent; e.g. the orientation and architecture of cytoskeletal structures, cilia and flagella, prions, proteins that replicate by changing the structure of normal proteins to match their own
3. chromatin marks, in which methyl or acetyl groups bind to DNA nucleotides or histones thereby altering gene expression patterns; e.g. Lcyc gene in Linaria vulgaris described below
4. RNA silencing, in which small RNA strands interfere (RNAi) with the transcription of DNA or translation of mRNA; known only from a few studies, mostly in Caenorhabditis elegans

Inheritance of epigenetic marks

Epigenetic variation may take one of four general forms. Others may yet be elucidated, but currently self-sustaining feedback loops, spatial templating, chromatin marking, and RNA-mediated pathways modify epigenes at the level of individual cells. Epigenetic variation within multicellular organisms may be endogenous, generated by cell–cell signaling (e.g. during cell differentiation early in development), or exogenous, a cellular response to environmental cues.

Removal vs. retention

In sexually reproducing organisms, much of the epigenetic modification within cells is reset during meiosis (e.g. marks at the FLC locus controlling plant vernalization), though some epigenetic responses have been shown to be conserved (e.g. transposon methylation in plants). Differential inheritance of epigenetic marks due to underlying maternal or paternal biases in removal or retention mechanisms may lead to the assignment of epigenetic causation to some parent of origin effects in animals and plants.

Reprogramming

In mammals, epigenetic marks are erased during two phases of the life cycle. Firstly just after fertilization and secondly, in the developing primordial germ cells, the precursors to future gametes. During fertilization the male and female gametes join in different cell cycle states and with different configuration of the genome. The epigenetic marks of the male are rapidly diluted. First, the protamines associated with male DNA are replaced with histones from the female's cytoplasm, most of which are acetylated due to either higher abundance of acetylated histones in the female's cytoplasm or through preferential binding of the male DNA to acetylated histones. Second, male DNA is systematically demethylated in many organisms, possibly through 5-hydroxymethylcytosine. However, some epigenetic marks, particularly maternal DNA methylation, can escape this reprogramming; leading to parental imprinting.

In the primordial germ cells (PGC) there is a more extensive erasure of epigenetic information. However, some rare sites can also evade erasure of DNA methylation. If epigenetic marks evade erasure during both zygotic and PGC reprogramming events, this could enable transgenerational epigenetic inheritance.

Recognition of the importance of epigenetic programming to the establishment and fixation of cell line identity during early embryogenesis has recently stimulated interest in artificial removal of epigenetic programming. Epigenetic manipulations may allow for restoration of totipotency in stem cells or cells more generally, thus generalizing regenerative medicine.

Retention

Cellular mechanisms may allow for co-transmission of some epigenetic marks. During replication, DNA polymerases working on the leading and lagging strands are coupled by the DNA processivity factor proliferating cell nuclear antigen (PCNA), which has also been implicated in patterning and strand crosstalk that allows for copy fidelity of epigenetic marks. Work on histone modification copy fidelity has remained in the model phase, but early efforts suggest that modifications of new histones are patterned on those of the old histones and that new and old histones randomly assort between the two daughter DNA strands. With respect to transfer to the next generation, many marks are removed as described above. Emerging studies are finding patterns of epigenetic conservation across generations. For instance, centromeric satellites resist demethylation. The mechanism responsible for this conservation is not known, though some evidence suggests that methylation of histones may contribute. Dysregulation of the promoter methylation timing associated with gene expression dysregulation in the embryo was also identified.

Decay

Whereas the mutation rate in a given 100 base gene may be 10⁻⁷ per generation, epigenes may "mutate" several times per generation or may be fixed for many generations. This raises the question: do changes in epigene frequencies constitute evolution? Rapidly decaying epigenetic effects on phenotypes (i.e. lasting less than three generations) may explain some of the residual variation in phenotypes after genotype and environment are accounted for. However, distinguishing these short-term effects from the effects of the maternal environment on early ontogeny remains a challenge.

Contribution to phenotypes

The relative importance of genetic and epigenetic inheritance is subject to debate. Though hundreds of examples of epigenetic modification of phenotypes have been published, few studies have been conducted outside of the laboratory setting. Therefore, the interactions of genes and epigenes with the environment cannot be inferred despite the central role of environment in natural selection. Experimental methodologies for manipulating epigenetic mechanisms are nascent (e.g.) and will need rigorous demonstration before studies explicitly testing the relative contributions of genotype, environment, and epigenotype are feasible.

In humans

A number of studies suggest the existence of transgenerational epigenetic inheritance in humans. These include those of the Dutch famine of 1944, wherein the offspring born during the famine were smaller than those born the year before the famine and the effects could last for two generations. Moreover, these offspring were found to have an increased risk of glucose intolerance in adulthood. Differential DNA methylation has been found in adult female offspring who had been exposed to famine in utero, but it is unknown whether these differences are present in their germline.

The Överkalix study noted sex-specific effects; a greater body mass index (BMI) at 9 years in sons, but not daughters, of fathers who began smoking early. The paternal grandfather's food supply was only linked to the mortality RR of grandsons and not granddaughters. The paternal grandmother's food supply was only associated with the granddaughters' mortality risk ratio. When the grandmother had a good food supply was associated with a twofold higher mortality (RR). This transgenerational inheritance was observed with exposure during the slow growth period (SGP). The SGP is the time before the start of puberty, when environmental factors have a larger impact on the body. The ancestors' SGP in this study, was set between the ages of 9-12 for boys and 8–10 years for girls. This occurred in the SGP of both grandparents, or during the gestation period/infant life of the grandmothers, but not during either grandparent's puberty. The father's poor food supply and the mother's good food supply were associated with a lower risk of cardiovascular death.

The loss of genetic expression which results in Prader–Willi syndrome or Angelman syndrome has in some cases been found to be caused by epigenetic changes (or "epimutations") on both the alleles, rather than involving any genetic mutation. In all 19 informative cases, the epimutations that, together with physiological imprinting and therefore silencing of the other allele, were causing these syndromes were localized on a chromosome with a specific parental and grandparental origin. Specifically, the paternally derived chromosome carried an abnormal maternal mark at the SNURF-SNRPN, and this abnormal mark was inherited from the paternal grandmother.

Similarly, epimutations on the MLH1 gene has been found in two individuals with a phenotype of hereditary nonpolyposis colorectal cancer, and without any frank MLH1 mutation which otherwise causes the disease. The same epimutations were also found on the spermatozoa of one of the individuals, indicating the potential to be transmitted to offspring.

A study has shown childhood abuse (defined in this study as "sexual contact, severe physical abuse and/or severe neglect") leads to epigenetic modifications of glucocorticoid receptor expression which play a role in HPA (hypothalamic-pituitary-adrenal) activity. Animal experiments have shown that epigenetic changes depend on mother-infant interactions after birth. In a recent study investigating correlations among maternal stress in pregnancy and methylation in teenagers and their mothers, it has been found that children of women who were abused during pregnancy were significantly more likely than others to have methylated glucocorticoid-receptor genes, which in turn change the response to stress, leading to a higher susceptibility to anxiety.

Effects on fitness

Epigenetic inheritance may only affect fitness if it predictably alters a trait under selection. Evidence has been forwarded that environmental stimuli are important agents in the alteration of epigenes. Ironically, Darwinian evolution may act on these neo-Lamarckian acquired characteristics as well as the cellular mechanisms producing them (e.g. methyltransferase genes). Epigenetic inheritance may confer a fitness benefit to organisms that deal with environmental changes at intermediate timescales. Short-cycling changes are likely to have DNA-encoded regulatory processes, as the probability of the offspring needing to respond to changes multiple times during their lifespans is high. On the other end, natural selection will act on populations experiencing changes on longer-cycling environmental changes. In these cases, if epigenetic priming of the next generation is deleterious to fitness over most of the interval (e.g. misinformation about the environment), these genotypes and epigenotypes will be lost. For intermediate time cycles, the probability of the offspring encountering a similar environment is sufficiently high without substantial selective pressure on individuals lacking a genetic architecture capable of responding to the environment. Naturally, the absolute lengths of short, intermediate, and long environmental cycles will depend on the trait, the length of epigenetic memory, and the generation time of the organism. Much of the interpretation of epigenetic fitness effects centers on the hypothesis that epigenes are important contributors to phenotypes, which remains to be resolved.

Deleterious effects

Inherited epigenetic marks may be important for regulating important components of fitness. In plants, for instance, the Lcyc gene in Linaria vulgaris controls the symmetry of the flower. Linnaeus first described radially symmetric mutants, which arises when Lcyc is heavily methylated. Given the importance of floral shape to pollinators, methylation of Lcyc homologues (e.g. CYCLOIDEA) may have deleterious effects on plant fitness. In animals, numerous studies have shown that inherited epigenetic marks can increase susceptibility to disease. Transgenerational epigenetic influences are also suggested to contribute to disease, especially cancer, in humans. Tumor methylation patterns in gene promotors have been shown to correlate positively with familial history of cancer. Furthermore, methylation of the MSH2 gene is correlated with early-onset colorectal and endometrial cancers.

Putatively adaptive effects

Experimentally demethylated seeds of the model organism Arabidopsis thaliana have significantly higher mortality, stunted growth, delayed flowering, and lower fruit set, indicating that epigenes may increase fitness. Furthermore, environmentally induced epigenetic responses to stress have been shown to be inherited and positively correlated with fitness. In animals, communal nesting changes mouse behavior increasing parental care regimes and social abilities that are hypothesized to increase offspring survival and access to resources (such as food and mates), respectively.

Macroevolutionary patterns

Inherited epigenetic effects on phenotypes have been documented in bacteria, protists, fungi, plants, and animals. Though no systematic study of epigenetic inheritance has been conducted (most focus on model organisms), there is preliminary evidence that this mode of inheritance is more important in plants than in animals. The early differentiation of animal germlines is likely to preclude epigenetic marking occurring later in development, while in plants and fungi somatic cells may be incorporated into the germ line.

Life history patterns may also contribute to the occurrence of epigenetic inheritance. Sessile organisms, those with low dispersal capability, and those with simple behavior may benefit most from conveying information to their offspring via epigenetic pathways. Geographic patterns may also emerge, where highly variable and highly conserved environments might host fewer species with important epigenetic inheritance.

Controversies in the history of inheritance

Humans have long recognized that traits of the parents are often seen in offspring. This insight led to the practical application of selective breeding of plants and animals, eventually leading to domestication, but did not address the central question of inheritance: how are these traits conserved between generations, and what causes variation?

Blending vs. particulate inheritance

Addressing these related questions, scientists during the time of the Enlightenment largely argued for the blending hypothesis, in which parental traits were homogenized in the offspring much like buckets of different colored paint being mixed together. Critics of Charles Darwin's On the Origin of Species, pointed out that under this scheme of inheritance, variation would quickly be swamped by the majority phenotype. In the paint bucket analogy, this would be seen by mixing two colors together and then mixing the resulting color with only one of the parent colors 20 times; the rare variant color would quickly fade.

Unknown to most of the European scientific community, a monk by the name of Gregor Mendel had resolved the question of how traits are conserved between generations through breeding experiments with pea plants. Charles Darwin thus did not know of Mendel's proposed "particulate inheritance" in which traits were not blended but passed to offspring in discrete units that we now call genes. Darwin came to reject the blending hypothesis even though his ideas and Mendel's were not unified until the 1930s, a period referred to as the Modern evolutionary synthesis.

Inheritance of innate vs. acquired characteristics

In his 1809 book, Philosophie Zoologique, Jean-Baptiste Lamarck recognized that each species experiences a unique set of challenges due to its form and environment. Thus, he proposed that the characters used most often would accumulate a "nervous fluid." Such acquired accumulations would then be transmitted to the individual’s offspring. In modern terms, a nervous fluid transmitted to offspring would be a form of epigenetic inheritance.

Lamarckism, as this body of thought became known, was the standard explanation for change in species over time when Charles Darwin and Alfred Russel Wallace co-proposed a theory of evolution by natural selection in 1859. Responding to Darwin and Wallace's theory, a revised neo-Lamarckism attracted a small following of biologists, though the Lamarckian zeal was quenched in large part due to Weismann’s famous experiment in which he cut off the tails of mice over several successive generations without having any effect on tail length. Thus the emergent consensus that acquired characteristics could not be inherited became canon.

Revision of evolutionary theory

Non-genetic variation and inheritance, however, proved to be quite common. Concurrent to the Modern Evolutionary Synthesis (unifying Mendelian genetics and natural selection), C. H. Waddington was working to unify developmental biology and genetics. In so doing, he coined the word "epigenetic" to represent the ordered differentiation of embryonic cells into functionally distinct cell types despite having identical primary structure of their DNA. Waddington’s epigenetics was sporadically discussed, becoming more of a catch-all for puzzling non-genetic heritable characters rather than advancing the body of inquiry. Consequently, the definition of Waddington’s word has itself evolved, broadening beyond the subset of developmentally signaled, inherited cell specialization.

Some scientists have questioned if inheritance compromises the foundation of the Modern Evolutionary Synthesis. Outlining the Central Dogma of Molecular Biology, Francis Crick succinctly stated, "DNA is held in a configuration by histone[s] so that it can act as a passive template for the simultaneous synthesis of RNA and protein[s]. None of the detailed 'information' is in the histone." (Italic added for emphasis.) However, he closes the article stating, "this scheme explains the majority of the present experimental results!" (Italic added for emphasis.) Indeed, the emergence of epigenetic inheritance (in addition to advances in the study of evolutionary-development, phenotypic plasticity, evolvability, and systems biology) has strained the current framework of the Modern Evolutionary Synthesis and prompted the re-examination of previously dismissed evolutionary mechanisms.

There has been much critical discussion of main stream evolutionary theory by Edward J Steele, Robyn A Lindley and colleagues, Fred Hoyle and N. Chandra Wickramasinghe, Yongsheng Liu Denis Noble, John Mattick and others that the logical inconsistencies as well as Lamarckian Inheritance effects involving direct DNA modifications, as well as the just described indirect, viz. epigenetic, transmissions, challenge conventional thinking in evolutionary biology and adjacent fields.